|
This thread is carried over and continued in the Current Elsmar Cove Forums 
|
|
|
This thread is carried over and continued in the Current Elsmar Cove Forums
|
The New Elsmar Cove Forums
|
 The New Elsmar Cove Forums
 Statistical Techniques and 6 Sigma
 High CPK and XR
|
 UBBFriend: Email This Page to Someone! |
next newest topic | next oldest topic |
| Author | Topic: High CPK and XR |
|
margherita Lurker (<10 Posts) Posts: 6 |
posted 26 October 1999 03:42 PM
   
I don't know how to I can combine high CPK (3, 4 or 5) with a set of XR charts. Too often the process uses (control limits) only a little part of the tollerance. It is in control but sometime the operator has to adjust it because there is an out of control signal or because the process changes, for any reason, its position. Too often it is difficult to understand this change because it is too small. Can anybody suggest me a more powerful SPC tool? Thanks IP: Logged |
|
Laura M Forum Contributor Posts: 299 |
posted 26 October 1999 04:15 PM
I've come across this in machining appliations and went to individuals and moving range charts (I-MR). Often times within subgroup variation was so small that it drove "artificially" narrow control limits. With R-bar being next to zero, the slightest mean shift makes the next subgroup go out when you are still well within tolerance range. Don't know if this is what you are experiencing or not, but my first suggestion would be to make sure you have rational subgroups. Laura IP: Logged |
|
Laura M Forum Contributor Posts: 299 |
posted 03 November 1999 06:36 PM
Have you tried this? Did it help? Anyone else out there have any other ideas? Just thought I'd bring this one to the top of the pile again to see if there are other ideas out there. IP: Logged |
|
Marc Smith Cheech Wizard Posts: 4119 |
posted 03 November 1999 06:38 PM
We need Don on this one. IP: Logged |
|
Don Winton Forum Contributor Posts: 498 |
posted 03 November 1999 08:50 PM
Well, I am back. My apologies for the delay.
I assume we are speaking of Xbar-R charts. What follows assumes so.
If this is the case, you would probably want to evaluate whether a control chart would add value at all. Too often, we tend to see the chart as a required item when in actuality it is not. Determine if the chart would add value then determine if the chart is useful. From the brief synopsis above, it sorta sounds you have a solution looking for a problem.
Sounds like you may want to chart the inputs to the process rather than the outputs. Just a thought.
There may be several. For example, you could use a variation on Laura Mâs suggestion. You could chart the process capability rather than the process using the X-mR chart, you could use Xbar charts on the inputs rather than the outputs or ignore charts altogether and concentrate on process capability as a metric of your process. The choice is up to you. At *** Dead Link Removed ***, there are several pages that offer control chart suggestions. Try there. But, as I stated above, be certain a chart is your best choice. Now, a Rant. It is a mistake to try to push an SPC apparatus where it neither adds value nor contributes to your productivity. Haphazardly using SPC usually causes more harm than good. This should be foremost on your mind. End Rant. Regards, IP: Logged |
All times are Eastern Standard Time (USA) | next newest topic | next oldest topic |
  |
Hop to: |
Your Input Into These Forums Is Appreciated! Thanks!
Main Site Search 
